Project Summary: Efforts to develop neuroprosthetic devices that restore function for patients who have lost vision, hearing, or somatic sensation typically aim to stimulate the sensory pathways of the nervous system in a manner that mimics their normal function closely. Yet the well-known cochlear implant for deaf patients has been successful even though the subject?s brain must adapt to the artificial stimulation, learning to interpret sounds and discriminate speech. Is biomimetic stimulation of sensory pathways the only way to provide neuroprosthetic inputs to the nervous system? In preliminary studies, we recently found that subjects could learn to interpret intracortical microstimulation (ICMS) delivered through 1 of 4 different electrodes in the premotor cortex (PM) as instructions to perform 1 of 4 arbitrarily associated movements. Even though ICMS in PM is not thought to evoke sensory percepts, subjects learned to use PM-ICMS instructions at currents and frequencies too low to evoke any muscle contraction or movement. Moreover, after the assignment of electrodes to movements was shuffled randomly subjects relearned the task, indicating that low-amplitude ICMS did not simply bias the subjects to perform specific movements, but instead evoked percepts or other experiences that the subjects could distinguish and learned to interpret as instructions. Here we propose to investigate whether subjects can experience, distinguish and learn to interpret low- amplitude ICMS delivered through different single electrodes in other select areas of the frontal and parietal association cortex that receive sensory information only indirectly. In Aim 1 we will examine 5 frontal areas: the supplementary motor area, the frontal eye field, the dorsolateral and ventrolateral prefrontal cortex, and the dorsal anterior cingulate cortex. In Aim 2 we will examine 5 parietal areas: area 5, the medial intraparietal area (parietal reach region), the lateral intraparietal area, the anterior intraparietal area, and area 7a. In Aim 3 we will determine whether subjects can learn to interpret ICMS in these frontal and parietal areas not only as instructions, but alternatively as feedback. The results of this work will expand the territory available for delivering artificial inputs to the nervous system substantially. Neuroprosthetic devices then may be developed that use such inputs to help patients affected by a wide variety of diseases of the central and peripheral nervous system including visual loss, sensory neuronopathies, stroke, and multiple sclerosis.